How to Edit Podcast Audio: A Workflow That Stays Consistent

Most podcast edits don't go wrong in a dramatic way. Nobody writes in about one bad episode. What listeners notice is that this week's show is louder than last week's, that the guest sounds a room away from the host, or that the intro music swamps your first sentence. Editing podcast audio is less about clever effects and more about doing the same handful of jobs, in the same order, every time you publish.

Why episodes drift in volume from week to week

Here's the habit behind most of that inconsistency: editing each episode by ear. You open the file, nudge the levels until it sounds fine on the headphones you're wearing that day, and export. The next week you do the same with a different guest, a different room and maybe a different pair of headphones. Each episode sounds acceptable on its own. Played back to back in someone's feed, they don't match.

Auphonic's own page describes loudness specifications as a way to define three things for a file: a target loudness, a true peak limit and a maximum loudness range, so that files stay consistent with each other. That's the fix in a sentence. You choose one target once, save it as a preset or write it in a note next to your episode template, and apply it to every episode without deciding again.

The true peak limit matters as much as the target. It's the ceiling that stops the loudest moments, a laugh or a hard consonant close to the microphone, from clipping. Set it alongside the loudness target and check it before every export.

I'm not going to hand you a single number to copy here, because the right target depends on where your show is heard and what your podcast host asks for. What matters is that you pick one and stick to it. A setting you apply to every episode beats a perfect setting you only remember half the time.

Six jobs, in the order that saves rework

1. Protect the raw files. Record each speaker on a clean track if you can, and copy the originals into a folder you never edit. Riverside records solo or with guests, which makes separate tracks easier to get. Edits are far easier to undo when the untouched audio is safe.
2. Cut first. Remove false starts, long pauses, coughs and filler words, listening through the whole episode once as you go. Descript lets you do this by editing the transcript, so deleting a line of text deletes it from the audio. Tight pacing is the edit listeners notice most.
3. Clean and balance. Reduce steady background noise and even out the volume between speakers so a quiet guest is as clear as you are. Auphonic lists noise and reverb reduction and an Intelligent Leveler for this job. Level differences are what send listeners reaching for the volume button.
4. Apply your loudness target. Use the same target on every episode, then check the true peak so the file doesn't clip. Auphonic's loudness specifications are built for exactly this step.
5. Add music under the voice. Bring intro and outro music in at a lower level than speech, then listen on headphones and on a phone speaker. Alitu's audio editor is made for podcasting and suits this kind of assembly. Music that competes with speech is a common complaint.
6. Export and upload. Save in the format your podcast host requests, name the file clearly, and upload it with a transcript or show notes. Audacity exports to formats such as MP3, FLAC and Ogg. A correct export avoids re-uploads and rejected files.

Cut before you process, never the other way round

The order above isn't arbitrary. If you clean and level the whole recording first and then delete ten minutes of it, you've spent processing time on audio nobody will hear, and the cuts can change how the remaining material sits together. Cutting first means every later step works on the finished running order.

The same logic applies to music. Add it after the voice is levelled, otherwise you're balancing music against a voice track that's about to change. And keep the loudness step near the end, so the reading you check is the one the listener gets.

There's one more reason to leave the raw files alone. Any clean-up can be pushed too far, and heavy processing can make a voice sound unnatural. If you've kept the original, you can go back and try a lighter setting. If you've edited over it, you're stuck with the result.

Five editors and the job each one handles

Descript

A text-based audio and video editor with multitrack editing. Its page lists Studio Sound, Remove Filler Words, Remove Retakes and Edit for Clarity, and the idea that sets it apart is that you edit the recording by editing its transcript.

Auphonic

Automatic audio post-production. Its page lists noise and reverb reduction, the Intelligent Leveler, cutting of filler words, coughs and silence, and the loudness specifications that keep files consistent from one episode to the next.

Riverside

A recording and editing platform. You record solo or with guests, then edit with a text-based editor and AI; its page lists Magic Audio and publishing to Spotify and Apple, so recording, editing and publishing can sit in one flow.

Alitu

A podcast maker aimed at people who don't want to become audio engineers. Its page says it removes background noises, evens out your voice and catches every 'uhm', with an audio editor made for podcasting.

Audacity

A free, open source audio editor and recorder that runs on Windows, macOS and Linux (free and open source as of Nov 2026), with plugin support. If you'd like a walkthrough for one of the fiddlier jobs, this guide to removing background noise one pass at a time starts from the same raw file you've just protected.

Automatic clean-up or hands-on control?

There are two broad ways to work. Automatic tools take a file, apply noise reduction, levelling and loudness in one go, and hand back something ready to publish. Editors that give you a timeline or a transcript let you decide every cut yourself. Automatic is quicker and more consistent; doing it by hand gives you control over pacing and the moments that make an episode feel human.

Many shows combine the two: cut by hand, then pass the edited file through an automatic tool for the technical clean-up and loudness. That split plays to each approach's strengths. To compare more options side by side, there's a wider roundup of podcast editing software for every skill level, and a shortlist of free audio editors if you'd rather start without signing up to anything.

A final listening check before you publish

• Headphones: listen for clicks, breaths and cuts that land in the middle of a word.
• Phone speaker: check the music isn't drowning out your voice.
• The first and last minute: make sure nothing was trimmed by mistake.
• Guest sections: confirm the quieter speaker is now as clear as the host.
• File name and format: match what your podcast host asks for.

My advice if you only change one habit

Write down one loudness target and one true peak limit, keep them where you'll see them, and apply them to every episode from now on. You don't need the fanciest editor to sound consistent. You need the same settings, the same order of jobs, and a final listen on a phone speaker before anything goes live. That's what makes a feed sound like one show rather than a pile of separate recordings.

Want your tool on this list? Featured spots and reviews here are paid partnerships. Here's how to partner with me.

Frequently asked questions

What is the best order for editing podcast audio?

Cut mistakes and dead air first, then balance levels and reduce noise, then set a loudness target, add music and export. Cutting first means you don't spend processing time on audio you're about to delete.

Can I edit a podcast for free?

Yes. Audacity's page describes a free, open source audio editor and recorder for Windows, macOS and Linux (as of Nov 2026).

How do I remove filler words from a podcast?

Descript's page lists Remove Filler Words, Auphonic's page lists cutting filler words, coughs and silence, and Alitu's page says it catches every 'uhm'.

Do I need to record separate tracks for each speaker?

It makes noise reduction and level balancing easier. Auphonic lists multitrack algorithms, and Riverside's page describes recording solo or with guests.

Should I use an automatic tool or edit by hand?

Automatic tools handle clean-up quickly, while timeline and transcript editors give you control over every cut. Many people combine both: cut by hand, then process automatically.